Solving fuzzy linear systems

In this paper, the main aim is to develop a method for solving an arbitrary general fuzzy linear system by using the embedding approach. Considering the existing and uniqueness of fuzzy solution to n × n linear fuzzy system is done. Numerical examples are presented to illustrate the proposed model.     

An accelerated randomized Kaczmarz method via low-rank approximation

The Kaczmarz method for finding the solution to an overdetermined consistent system of linear equation Ax=b(A∈R^m×n) is an iterative algorithm that has found many applications ranging from computer tomography to digital signal processing. Recently, Strohmer and Vershynin proposed randomized Kaczmarz, and proved its exponential convergence. In this paper, motivated by idea of precondition, we present a modified version of the randomized Kaczmarz method where an orthogonal matrix was multiplied to both sides of the equation Ax=b, and the orthogonal matrix is obtained by low-rank approximation. Our approach fits the problem when m is huge and m?n. Theoretically, we improve the convergence rate of the randomized Kaczmarz method. The numerical results show that our approach is faster than the standard randomized Kaczmarz.     

An important property of non-minimum-phase multiple-input-multiple-output feedback systems

The non-minimum-phase (NMP( property is easily determined from the requirement that the plant input is bounded. In the single-input-single-output (SISO) system, a right-half-plane (RHP) plant zero at s = b constrains the system transfer function to have a zero at b. Also, the available feedback benefits are significantly restricted. The n × n multiple-input-multiple-output (MIMO) system is NMP if the plant determinant δhas any RHP zeros, say at plant transfer matrix and T = [t_ij is the closed-loop system transfer matrix. It has been thought that all n²t_ij (and the n₂ plant disturbance response function rfj), must suffer from the NMP liability in their feedback properties. It is shown that only one row of need so suffer, with a any fixed integer in [1, n].The remaining n(n — 1) elements can be completely free of the NMP liability. A mathematically rigorous synthesis technique previously developed for MP systems is shown to be well suited for precise numerical design for such NMP MIMO plants with significant uncertainties. In this technique, the MIMO design problem is converted into a number of equivalent SISO problems. An example involving disturbance attenuation in a highly uncertain 2×2 NMP plant is included.     

Solving fully fuzzy polynomials using feed-back neural networks

Recently, there has been a considerable amount of interest and practice in solving many problems of several applied fields by fuzzy polynomials. In this paper, we have designed an artificial fuzzified feed-back neural network. With this design, we are able to find a solution of fully fuzzy polynomial with degree n. This neural network can get a fuzzy vector as an input, and calculates its corresponding fuzzy output. It is clear that the input–output relation for each unit of fuzzy neural network is defined by the extension principle of Zadeh. In this work, a cost function is also defined for the level sets of fuzzy output and fuzzy target. Next a learning algorithm based on the gradient descent method will be defined that can adjust the fuzzy connection weights. Finally, our approach is illustrated by computer simulations on numerical examples. It is worthwhile to mention that application of this method in fluid mechanics has been shown by an example.     

Numerical solution of fully fuzzy linear systems by fuzzy neural network

In this paper, a new hybrid method based on fuzzy neural network (FNN) for approximate solution of fuzzy linear systems of the form Ax=d,Ax=d, where AA is a square matrix of fuzzy coefficients, xx and dd are fuzzy number vectors, is presented. Here a neural network is considered as a part of a large field called neural computing or soft computing. Moreover, in order to find the approximate solution of an n×nn\times n system of fuzzy linear equations that supposedly has a unique fuzzy solution, a simple algorithm from the cost function of the FNN is proposed. Finally, we illustrate our approach by some numerical examples.     

The accelerated overrelaxation quadrant interlocking iterative method

In this paper we use a new splitting of the matrix A of the linear system A x = b introduced in [1] and we present a new version of the AOR method, more suitable for parallel processing, which involves explicit evaluation of 2 × 2 blocks.

We also obtain several convergence conditions for this new method, when the matrix A of (1.1) belongs to different classes of matrices. Some results, given in [1], are also improved and generalised.     

Efficient Exact Evaluation of Signs of Determinants

This paper presents a theoretical and experimental study on two different methods to evaluate the sign of a determinant with integer entries. The first one is a method based on the Gram—Schmidt orthogonalization process which has been proposed by Clarkson [Cl]. We review his algorithm and propose a variant of his method, for which we give a complete analysis. The second method is an extension to n × n determinants of the ABDPY method [ABD+2] which works only for 2 × 2 and 3 × 3 determinants. Both methods compute the sign of an n× n determinant whose entries are integers on b bits, by using exact arithmetic on only b +O(n) bits. Furthermore, both methods are adaptive, dealing quickly with easy cases and resorting to full-length computation only for null determinants. Received December 30, 1996; revised September 16, 1998.     

Further results on controllability and the linear equation Ax = b

The linear equation Ax = b, with A an n × n matrix and b an n × l matrix over a unique factorization domain R, is related to the controllability submodule U of the pair (A, b). It is shown that the above equation has a solution lying in V if, and only if, A is unimodular as an operator on U. An example is given of a matrix which is unimodular as an operator on the controllability submodule, but not as an operator on Rⁿ and sparseness of this occurrence is discussed.     

Sequential Maximum Correntropy Kalman Filtering

This paper explores a linear state estimation problem in non‐Gaussian setting and suggests a computationally simple estimator based on the maximum correntropy criterion Kalman filter (MCC‐KF). The first MCC‐KF method was developed in Joseph stabilized form. It requires two n × n and one m × m matrix inversions, where n is a dimension of unknown dynamic state to be estimated, and m is a dimension of available measurement vector. Therefore, the estimator becomes impractical when the system dimensions increase. Our previous work has suggested an improved MCC‐KF estimator (IMCC‐KF) and its factored‐from (square‐root) implementations that enhance the MCC‐KF estimation quality and numerical robustness against roundoff errors. However, the proposed IMCC‐KF and its square‐root implementations still require the m × m matrix inversion in each iteration step of the filter. For numerical stability and computational complexity reasons it is preferable to avoid the matrix inversion operation. In this paper, we suggest a new IMCC‐KF algorithm that is more accurate and computationally cheaper than the original MCC‐KF and previously suggested IMCC‐KF. Furthermore, compared with stable square‐root algorithms, the new method is also accurate, but less computationally expensive. The results of numerical experiments substantiate the mentioned properties of the new estimator on numerical examples.     

Quantitative design method for MIMO uncertain plants to achieve prescribed diagonal dominant closed-loop minimum-phase tolerances

A quantitative design method for multi-input multi-output linear time-invariant feedback systems for plants with large uncertainty has been presented by Horowitz ( 1982), and by Yaniv and Horowitz ( 1986). This design method is developed here to guarantee minimum-phase closed-loop diagonal elements for systems with basically non-interacting (Horowitz and Loecher 1981) off-diagonal closed-loop tolerances. The advantage of this design is that with minimum-phase transfer functions, a very important class of time-domain specifications can be translated to the frequency domain, as shown by Krishman and Cruickshanks ( 1977) and by Horowitz ( 1976). The attractive properties of this design method are: (a) the problem is reduced to a successive single-loop design with no interaction between the loops, and no iterations are necessary; (b) the technique can be applied to all n × n plants P with P^?1 having no poles in the right-half plane, and satisfying some conditions described in § 5; (c) the procedure is interactive with n steps for an n × n MIMO plant, and in each step, one of the elements of the diagonal feedback compensation and one row of the prefilter matrix are designed.     

A class of iterative methods for computing polar decomposition

In this article, there is offered a parametric class of iterative methods for computing the polar decomposition of a matrix. Each iteration of this class needs only one scalar-by-matrix and three matrix-by-matrix multiplications. It is no use computing inversion, so no numerical problems can be created because of ill-conditioning. Some available methods can be included in this class by choosing a suitable value for the parameter. There are obtained conditions under which this class is always quadratically convergent. The numerical comparison performed among six quadratically convergent methods for computing polar decomposition, and a special method of this class, chosen based on a specific value for the parameter, shows that the number of iterations of the special method is considerably near that of a cubically convergent Halley's method. Ten n×n matrices with n=5, 10, 20, 50, 100 were chosen to make this comparison.     

A method for solving nth order fuzzy linear differential equations

In this paper, an analytic method (eigenvalue–eigenvector method) for solving nth order fuzzy differential equations with fuzzy initial conditions is considered. In this method, three cases are introduced, in each case, it is shown that the solution of differential equation is a fuzzy number. In addition, the method is illustrated by solving several numerical examples.     

A NOTE ON “DECOMPOSITION PROBLEM OF FUZZY RELATIONS”

The decomposition problem of a fuzzy relation R ∈ F(X × X) can be stated as: “Given a fuzzy relation R∈F(X ×.X), to determine whether there exists a fuzzy relation Z∈f(X × X) such that R = Z [Odot] Z, where X is a finite set and “[Odot]” is the max-min composition of two fuzzy relations.” In particular, if R is a Boolean matrix, then this problem becomes to find the square root of a Boolean matrix, which is a well-known unsolved problem. In 1985, Di Nola et al. (A. Di Nola, S. Sessa and W. Pedrycz, Int J. General Systems,. 10, 1985, 123?133) had solved it in theory, and proposed a numerical algorithm, illustrated by a flowchart In this note, we first point out that the flowchart proposed by Di Nola et al. is in error and give a correct flowchart. Then we give a numerical example, which is also a counterexample of the flowchart given by Di Nola el al., to explain our flowchart.     

Approximate clustering in very large relational data

Different extensions of fuzzy c‐means (FCM) clustering have been developed to approximate FCM clustering in very large (unloadable) image (eFFCM) and object vector (geFFCM) data. Both extensions share three phases: (1) progressive sampling of the VL data, terminated when a sample passes a statistical goodness of fit test; (2) clustering with (literal or exact) FCM; and (3) noniterative extension of the literal clusters to the remainder of the data set. This article presents a comparable method for the remaining case of interest, namely, clustering in VL relational data. We will propose and discuss each of the four phases of eNERF and our algorithm for this last case: (1) finding distinguished features that monitor progressive sampling, (2) progressively sampling a square N × N relation matrix R_N until an n × n sample relation R_n passes a statistical test, (3) clustering R_n with literal non‐Euclidean relational fuzzy c‐means, and (4) extending the clusters in R_n to the remainder of the relational data. The extension phase in this third case is not as straightforward as it was in the image and object data cases, but our numerical examples suggest that eNERF has the same approximation qualities that eFFCM and geFFCM do. © 2006 Wiley Periodicals, Inc. Int J Int Syst 21: 817–841, 2006.     

Activation and Defuzzification Methods for Fuzzy Rule-Based Systems

In this paper, a Selective Inference Engine (SIE) is first proposed. SIE predicts the rules that will be fired, based on an anticipated location procedure, and then performs the inference calculations only on the latter. This anticipated location is based on the projection of the input data on the conditional space of the fuzzy system and the delimitation of the excited region. Then, the fired rules can be aggregated using the appropriate scheme. In the second part of this work, we propose new defuzzification methods which take into account the consequent membership function shapes.     

On the new solutions for a fully fuzzy linear system

In this paper, we propose a new method to present a fuzzy trapezoidal solution, namely “suitable solution”, for a fully fuzzy linear system (FFLS) based on solving two fully interval linear systems (FILSs) that are 1-cut and 0-cut of the related FILS. After some manipulations, two FILSs are transformed to 2n crisp linear equations and 4n crisp linear nonequations and n crisp nonlinear equations. Then, we propose a nonlinear programming problem (NLP) to computing simultaneous (synchronic) equations and nonequations. Moreover, we define two other new solutions namely, “fuzzy surrounding solution” and “fuzzy peripheral solution” for an FFLS. It is shown that the fuzzy surrounding solution is placed in a tolerable fuzzy solution set and the fuzzy peripheral solution is placed in a controllable fuzzy solution set. Finally, some numerical examples are given to illustrate the ability of the proposed methods.     

New gauss–seidel like block iterative methods to solve discrete boundary value problems

The concept of new Gauss–Seidel like iterative methods, which was introduced in [3], will be extended so as to obtain a class of convergent Gauss–Seidel like block iterative methods to solve linear matrix equations Ax=b with an M-Matrix A. New block iterative methods will be applied to finite difference approximations of the Laplace's equation on a square (“model problem” [8]) which surpass even the block successive overrelaxation iterative method with optimum relaxation factor in this example.     

An effective approach to the optimal-control problem for time-varying linear systems via Taylor series

The Taylor series is used for the solution of the optimal-control problem for time-varying linear systems. Instead of solving the state transition matrix from the state equation with a terminal condition, the present approach first transforms the terminal condition into an initial condition, and then solves the initial-value problem to find the transition matrix. This approach leads lo a recursive algebraic formulation for the transition matrix, and only an inverse matrix of small dimension 2n × 2n appears in this formulation. Thus a closed-loop control law is obtained without solving the non-linear Riccati equation, and the matrix to be inverted has only small dimension 2n × 2n. The present approach is of great interest because of its simplicity and numerical stability.     

Iterative solution of a class of linear equations with application to reconstruction of three-dimensional object arrays†

An iterative algorithm baaed on probabilistic estimation is described for obtaining the minimum-norm solution of a very large, consistent, linear system of equations AX = g where A is an (m × n) matrix with non-negative elements, x and g are respectively (n × 1) and (m × 1) vectors with positive components.

This algorithm will find application in the reconstruction of three-dimensional object arrays from projections and in several other areas.     

New results on static output feedback H ∞ control for fuzzy singularly perturbed systems: a linear matrix inequality approach

This paper presents the novel approaches of designing robust fuzzy static output feedback H _∞ controller for a class of nonlinear singularly perturbed systems. Specifically, the considered system is approximated by a fuzzy singularly perturbed model. With the use of linear matrix inequality (LMI) methods, two methods are provided to design fuzzy static output feedback H _∞ controllers. The resulted controllers can guarantee that the closed‐loop systems are asymptotically stable and satisfy H _∞ performances for sufficiently small ?. In contrast to the existing results, the proposed approaches have two advantages: (i) the gains of controller are solved directly by a set of ?‐independent LMIs, and therefore, the problem of selecting the initial values in iterative LMIs algorithm can be avoided, and (ii) the smaller control input efforts are needed. The given methods are easy to implement and can be applied to both standard and nonstandard nonlinear singularly perturbed systems. Two numerical examples are provided to illustrate the effectiveness of the developed methods. Copyright © 2012 John Wiley & Sons, Ltd.